Earth reflectivity from Deep Space Climate Observatory (DSCOVR) Earth Polychromatic Camera (EPIC)
MetadataShow full item record
Citation (published version)Wanjuan Song, Juri Knjazihhin, Guoyong Wen, Alexander Marshak, Guangjian Yan, Xihan Mu, Taejin Park, Chi Chen, Baodong Xu, Ranga Myneni. 2017. "Earth Reflectivity from Deep Space Climate Observatory (DSCOVR) Earth Polychromatic Camera (EPIC)." 2017 AGU Fall Meeting, New Orleans, Louisiana.
Earth reflectivity, which is also specified as Earth albedo or Earth reflectance, is defined as the fraction of incident solar radiation reflected back to space at the top of the atmosphere. It is a key climate parameter that describes climate forcing and associated response of the climate system. Satellite is one of the most efficient ways to measure earth reflectivity. Conventional polar orbit and geostationary satellites observe the Earth at a specific local solar time or monitor only a specific area of the Earth. For the first time, the NASA’s Earth Polychromatic Imaging Camera (EPIC) onboard NOAA’s Deep Space Climate Observatory (DSCOVR) collects simultaneously radiance data of the entire sunlit earth at 8 km resolution at nadir every 65 to 110 min. It provides reflectivity images in backscattering direction with the scattering angle between 168º and 176º at 10 narrow spectral bands in ultraviolet, visible, and near-Infrared (NIR) wavelengths. We estimate the Earth reflectivity using DSCOVR EPIC observations and analyze errors in Earth reflectivity due to sampling strategy of polar orbit Terra/Aqua MODIS and geostationary Goddard Earth Observing System-R series missions. We also provide estimates of contributions from ocean, clouds, land and vegetation to the Earth reflectivity. Graphic abstract shows enhanced RGB EPIC images of the Earth taken on July-24-2016 at 7:04GMT and 15:48 GMT. Parallel lines depict a 2330 km wide Aqua MODIS swath. The plot shows diurnal courses of mean Earth reflectance over the Aqua swath (triangles) and the entire image (circles). In this example the relative difference between the mean reflectances is +34% at 7:04GMT and -16% at 15:48 GMT. Corresponding daily averages are 0.256 (0.044) and 0.231 (0.025). The relative precision estimated as root mean square relative error is 17.9% in this example.
Poster presented at 2017 AGU Fall Meeting, New Orleans, Louisiana. POSTER ID: A33D-2387